Description

Senior LLM Engineer | 3+ years | Switzerland | Regulated sector | Python & Azure OpenAI; build and scale production-grade RAG and agentic AI systems where performance, governance, auditability, and reliability are as important as model capability. About the role: We're hiring an LLM Engineer ; You'll design, build, and productionize LLM-powered systems from document and process automation to agentic workflows, inside an environment where governance, auditability, and reliability matter as much as capability. This is the right role for someone who has taken generative AI use cases from proof-of-concept to industrialized, monitored production systems, and who enjoys working close to the model while still thinking like a software engineer. Responsibilities: • Design, build, and deploy LLM-powered applications — including RAG and agentic workflows — from prototype through to production • Fine-tune, prompt-engineer, and evaluate LLMs for accuracy, latency, and cost across business use cases • Build and maintain retrieval-augmented and agentic architectures, including vector stores and orchestration frameworks (e.g. LangChain, LangGraph) • Implement safety measures — content filtering, bias detection, and guardrails — aligned to the organization's responsible-AI and governance requirements • Optimize inference performance, latency, and cost for high-volume, production workloads • Monitor deployed systems in production, track drift and performance degradation, and ship improvements • Collaborate with data engineering, MLOps, and platform teams on data pipelines, feature stores, and CI/CD for model deployment • Document architecture, evaluation results, and change history to support model risk, audit, and regulatory requirements Experience: • 3+ years of hands-on experience building and deploying ML/LLM systems in production, ideally inside a regulated environment • Proven track record taking generative AI proof-of-concepts into hardened, production-grade systems • Experience with evaluation frameworks, agent tooling, and RAG pipelines • Prior exposure to financial services, healthcare, or another regulated sector is a plus, not a requirement Technical fluency: • Strong Python engineering skills and solid software engineering fundamentals (testing, version control, CI/CD) • Good understanding of transformer architectures and practical experience with fine-tuning and prompt-optimization techniques • Hands-on experience with the Azure AI stack: Azure OpenAI, Azure AI Foundry, and Azure Machine Learning • Experience with vector databases and RAG/agent frameworks (LangChain, LangGraph, or equivalent) • Familiarity with MLOps/LLMOps practices — model deployment, monitoring, feature stores, and CI/CD (Azure DevOps or GitHub Actions) • Comfortable with SQL, Spark, Docker, and cloud data platforms (Azure Data Lake, Databricks, or Microsoft Fabric) Ways of working: • Agile delivery discipline, comfortable working inside cross-functional squads with data science, MLOps, and business analysis • Pragmatic engineering mindset that balances speed, cost, and regulatory constraints • Strong documentation habits to support governance, model risk, and audit needs Qualifications: • University degree in computer science, machine learning, or a related quantitative field (Master's a plus) • Fluent English and French; • Eligibility to work in Switzerland Your data: By submitting your resume, you agree to the retention and use of your personal data by TSG for recruitment purposes, including sharing with our clients in the context of your application. The identity and sector of the client will be disclosed to shortlisted candidates ahead of interview.